Perceive, Attend, and Drive: Learning Spatial Attention for Safe Self-Driving

Understand

In this paper, we propose an end-to-end self-driving network featuring a sparse attention module that learns to automatically attend to important regions of the input.

• The attention module specifically targets motion planning, whereas prior literature only applied attention in perception tasks.
• Learning an attention mask directly targeted for motion planning significantly improves the planner safety by performing more focused computation.
• Furthermore, visualizing the attention improves interpretability of end-to-end self-driving.